Effect of Vitamin D Supplementation on Recurrent Wheezing in Black Infants Who Were Born Preterm: The D-Wheeze Randomized Clinical Trial

Key Points

Question  Does sustained vitamin D supplementation until 6 months of age affect recurrent wheezing in black infants born preterm?

Findings  In this randomized clinical trial of 300 infants, sustained supplementation with 400 IU/d of vitamin D until 6 months of age, compared with a diet-limited approach, resulted in a likelihood of recurrent wheezing at 12 months of 31.1% vs 41.8%, respectively, a statistically significant difference.

Meaning  Among black infants born preterm, sustained supplementation with vitamin D, compared with diet-limited supplementation, resulted in a reduced risk of recurrent wheeze by 12 months’ adjusted age.

Importance  Black infants born preterm face high rates of recurrent wheezing throughout infancy. Vitamin D supplementation has the potential to positively or negatively affect wheezing through modulation of the pulmonary and immune systems.

Objective  To assess the effectiveness of 2 vitamin D dosing strategies in preventing recurrent wheezing.

Design, Setting, and Participants  A randomized clinical trial enrolled 300 black infants born at 28 to 36 weeks’ gestation between January 2013 and January 2016 at 4 sites in the United States, and followed them up through March 2017. Randomization was stratified by site and maternal milk exposure.

Interventions  Patients were enrolled prior to discharge from the neonatal intensive care unit or newborn nursery and received open-label multivitamin until they were consuming 200 IU/d of cholecalciferol from formula or fortifier added to human milk, after which they received either 400 IU/d of cholecalciferol until 6 months of age adjusted for prematurity (sustained supplementation) or placebo (diet-limited supplementation). One-hundred fifty three infants were randomized to the sustained group, and 147 were randomized to the diet-limited group.

Main Outcomes and Measures  Recurrent wheezing by 12 months’ adjusted age was the primary outcome.

Results  Among 300 patients who were randomized (mean gestational age, 33 weeks; median birth weight, 1.9 kg), 277 (92.3%) completed the trial. Recurrent wheezing was experienced by 31.1% of infants in the sustained supplementation group and 41.8% of infants in the diet-limited supplementation group (difference, −10.7% [95% CI, −27.4% to −2.9%]; relative risk, 0.66 [95% CI, 0.47 to 0.94]). Upper and lower respiratory tract infections were among the most commonly reported adverse events. Upper respiratory infections were experienced by 84 of 153 infants (54.9%) in the sustained group and 83 of 147 infants (56.5%) in the diet-limited group (difference, −1.6% [95% CI, −17.1% to 7.0%]). Lower respiratory infections were experienced by 33 of 153 infants (21.6%) in the sustained group and 37 of 147 infants (25.2%) in the diet-limited group (difference, −3.6% [95% CI, −16.4% to 4.4%]).

Conclusions and Relevance  Among black infants born preterm, sustained supplementation with vitamin D, compared with diet-limited supplementation, resulted in a reduced risk of recurrent wheezing by 12 months’ adjusted age. Future research is needed to better understand the mechanisms and longer-term effects of vitamin D supplementation on wheezing in children born preterm.

Trial Registration  ClinicalTrials.gov Identifier: NCT01601847

Quiz Ref IDWheezing is a common complication of prematurity. Early environmental and iatrogenic exposures may perturb the development of the lung, airway, or immune system and lead to recurrent wheezing. Substantial evidence suggests vitamin D is important in multiple pathways relevant to the development of wheezing in young children, including regulating inflammation, response to pathogens, lung and airway development, and propensity to allergic disease.^1-9 Preterm infants, with developmentally immature pulmonary and immune systems perturbed by preterm birth, may be particularly vulnerable to any positive and negative effects of vitamin D. Black infants experience higher rates of both prematurity and prematurity-associated wheezing, and previous epidemiological work suggests that race may affect the relationship between vitamin D exposure and wheezing, leading to the design of the present study focusing exclusively on black infants.^10,11

The Wheezing in Black Preterm Infants: Impact of Vitamin D Supplementation Strategy (D-Wheeze) Study assessed the effectiveness of 2 vitamin D supplementation strategies in black infants born preterm: (1) sustained supplementation with 400 IU/d of cholecalciferol until 6 months of age adjusted for prematurity, or (2) cessation of supplementation when the infant was taking 200 IU/d or more from diet. This trial focused on clinical outcomes as well as surrogate markers of pulmonary health, allergy, and bone health.

This study was approved by the institutional review board at each participating institution. The protocol is available in Supplement 1. Written informed consent was obtained from the infants’ parents prior to drawing any eligibility laboratories that were not obtained as part of clinical care and prior to randomization of eligible infants.

We conducted a masked placebo-controlled randomized clinical trial of the effectiveness of 2 vitamin D supplementation strategies in black infants born at 28^0/7 to 36^6/7 weeks’ gestational age (GA). Patients were enrolled between January 2013 and January 2016; the follow-up window closed March 2017. Infants were recruited from University Hospitals Cleveland Medical Center (Cleveland, Ohio), MetroHealth Medical Center (Cleveland, Ohio), the Medical University of South Carolina (Charleston), and Montefiore Medical Center (Bronx, New York).

Infants were randomized to receive 400 IU of cholecalciferol daily until 6 months’ adjusted age (sustained group) or until they were taking at least 200 IU/d of vitamin D from formula or human milk fortifier (diet-limited group). Patients received open-label multivitamin until they were consuming 200 IU/d of cholecalciferol from formula or human milk fortifier, after which they received masked study drug (liquid cholecalciferol or a placebo, dispensed in an amber bottle). Parents, clinical caregivers, and study staff with patient contact were masked. Exclusively breastfed infants received open-label multivitamin as long as they were exclusively breastfed, up to 12 months’ adjusted age. The study protocol did not dictate what infants were fed.

Study visits occurred at 3, 6, 9, and 12 months of age, adjusted for degree of prematurity, in the infant’s home and in the clinic. A monthly diet questionnaire was completed until 6 months’ adjusted age (or longer if exclusively breastfed) to assess 25-hydroxyvitamin D (25[OH]D) intake.

Quiz Ref IDInfants were eligible for enrollment if they were 28^0/7 to 36^6/7 weeks’ GA at birth, the family identified the child as black or African American, they received 28 days or less of supplemental oxygen, were admitted to a participating nursery as a neonate, were 40^6/7 weeks’ adjusted GA or younger at enrollment, and lived within a predefined geographic area at each site. Infant race was considered “black” or “African American” if the parents affirmed that they believed one of those descriptors applied to their infant.

Infants were not eligible if they were diagnosed as having bronchopulmonary dysplasia; had a preexisting diagnosis of moderate to severe osteopenia of prematurity and/or alkaline phosphatase level greater than 700 U/L (to convert to μkat/L, multiply by 0.0167); a history of fracture; a history of gastrointestinal surgery including for necrotizing enterocolitis, known gastrointestinal malabsorption, a major congenital anomaly, a congenital pulmonary or airway disorder, documented wheezing, or stridor prior to enrollment; previous vitamin D supplementation with more than 400 IU/d; or the family planned to move out of the region. Infants were also ineligible if their serum phosphorus concentration was outside of the range of 4.0 to 9.5 mg/dL (to convert to mmol/L, multiply by 0.323) or serum calcium was outside of the range of 8.5 to 10.7 mg/dL (to convert to mmol/L, multiply by 0.25). A 25(OH)D concentration less than 10 ng/mL or greater than 80 ng/mL also made infants ineligible (to convert to nmol/L, multiply by 2.496). Eligible ranges for laboratory values were modified in the first year of enrollment (eMethods in Supplement 2). Serum bone markers of health and 25(OH)D levels were processed at each site’s clinical chemistry laboratory.

Adherence was estimated by weighing bottles of either open-label multivitamins or masked study drug before they were dispensed and after they were returned. Ideal adherence was defined as a weight difference averaging approximately 1 g/d.

The trial was monitored by a National Institutes of Health–convened data and safety monitoring board. A medical monitor reviewed certain categories of adverse events and laboratory values out of a predetermined range (eMethods in Supplement 2). She had independent authorization to either remove a child from the study or refer for further medical care. She was authorized to view group allocation as needed for safety, but did not access this information.

Infants were randomized with randomly permuted blocks, sizes 2 to 6, using computer-generated random numbers. Randomization was stratified by site and whether the infant had received any of their own mother’s milk within 24 hours of randomization. Twins and triplets were randomized together. Families, clinical caregivers, and study staff were blinded to assignment and block size.

Quiz Ref IDThe primary outcome was parent-reported recurrent wheezing (≥2 episodes of wheezing with or without an infection) by 12 months’ adjusted age, using a modified International Study of Asthma and Allergies in Childhood questionnaire.¹² These questions have been shown to perform well in infants born preterm.¹³ Recurrent wheezing was defined as 1 report of more than 1 wheezing episode since the last interview or reports of single wheezing episodes at multiple visits. The ascertainment of recurrent wheezing was standardized with other concurrent trials.^14,15

Secondary outcomes included parental report of upper and lower respiratory tract infections, pulmonary medications, hospitalization and emergency department visits, allergies, and eczema. These were assessed by questionnaire at each follow-up point. The modified Asthma Predictive Index was calculated, which yields a positive or negative prediction.¹⁶ Eosinophil counts were obtained at 12 months. Scoring Atopic Dermatitis skin examinations for eczema were completed at each follow-up point; Scoring Atopic Dermatitis scores range from 0 to 103; and a value of 25 or more indicates moderate to severe eczema.¹⁷ Circulating 25(OH)D3 concentrations were measured at the 3-, 6-, and 12-month visits, using the immunoassays that were standard of care at each site. Clinical recommendations for infants generally target levels 20 ng/mL or greater.^18,19

Serum markers of bone health (alkaline phosphatase, calcium, phosphorus) were also measured at the clinical chemistry laboratory at each site. A speed-of-sound measure of tibial bone density assessment was done at 12 months. Published norms have described the mean and 2 SDs below the mean for 1-year-old girls to be 3139 m/s and 2919 m/s, respectively, and for boys to be 3189 m/s and 2871 m/s, respectively.²⁰ Values less than 2 SDs below the mean were considered abnormal. A panel of allergen-specific IgE antibodies (ImmunoCAP Phadiatop Infant; Phadia AB) was obtained at 12 months.²¹ The panel included common food and inhalant allergies (egg, cow’s milk, peanut, shrimp, cat, dog, house dust mite, birch, timothy, ragweed, and wall pellitory). Results range from 0 to 100 k_A/L; a level of 0.35 k_A/L or greater is considered positive.

Based on a baseline rate of recurrent wheezing of 30%, a 2-sided significance level of .05, and a power of 80% to detect a relative risk of 1.6, a sample size of 115 infants per group was estimated. Allowing for 15% loss to follow-up and 20% of enrolled infants to be part of a twin pair, a sample size of 300 infants was planned. In the absence of a generally accepted minimal clinically important difference for studies of recurrent wheezing in infancy, a relative risk of 1.6 was chosen to be sufficiently large to be meaningful at face value and to be plausible based on prior studies.¹⁰

Standard summary statistics were used to describe baseline characteristics. For characteristics measured at the family level, such as maternal age, Pearson χ² tests or 2-sample t tests were used. For comparisons of patient-level characteristics, such as birth weight, we used generalized estimating equations (GEEs) with exchangeable working covariance to reflect within-family covariance. All statistical analyses were conducted using SPSS version 25 (IBM Corp) or SAS for Windows 9.4 (SAS Institute Inc).

Analyses were based on a modified intent-to-treat approach. Because the primary outcome was a composite of longitudinal points, and most participants who withdrew did so before 3 months (Figure 1), withdrawn participants were not included. The 4 participants who were lost to all follow-up were not included in the primary analysis. Type I error was assumed to be .05, except for the primary outcome. One interim analysis was performed. The significance level used at the interim analysis was .003, while the significance level used at the final analysis was .047.

For the primary outcome of recurrent wheezing, we used GEEs to account for the nonindependence of twins and triplets and to estimate the risk ratio and risk difference.^22,23 Two Poisson regression models, adjusted and unadjusted, were planned a priori, with the unadjusted model considered primary. Covariates in the secondary adjusted model included randomization strata (site and maternal milk at enrollment), GA, time in study, and baseline covariates associated with recurrent wheezing within a significance level of .10. Adjustment for unbalanced baseline characteristics not associated with recurrent wheezing was not planned because, if a covariate is not strongly related to an outcome but is imbalanced between groups, adjustment for an imbalanced covariate may dilute the effect of treatment.²⁴

A pattern mixture modeling approach was used to assess the sensitivity of the primary finding to missing data. This included 8 participants whose recurrent wheezing outcome could not be determined due to a missing 12-month visit and not having yet met the definition of recurrent wheezing. We considered a range of scenarios where the outcome was no longer independent of whether or not the data were missing (ie, not missing at random). This missing data approach is detailed in the eMethods in Supplement 2.

For secondary binary outcomes that were planned a priori, we fit unadjusted Poisson GEE regression models. For continuous outcomes, we used linear GEE models. Secondary outcomes were also analyzed by follow-up point. Corresponding 95% CIs of risk and mean difference were generated from these models.

Post hoc analyses assessed the role of baseline maternal milk status and sex on treatment outcomes through modeling of interaction with randomization group. In addition, a post hoc analysis of the primary outcome was conducted, which added baseline unbalanced covariates (P < .05) to the adjusted model.

Of the 300 infants enrolled in the study, 18 withdrew from the study and 1 died while cosleeping (Figure 1). Follow-up rates of surviving nonwithdrawn infants at the 3-, 6-, 9-, and 12-month visits were 97.9%, 96.5%, 95.0%, and 94.0%, respectively. Due to missing 12-month visits in infants who had not yet met criteria for recurrent wheezing, we were unable to determine recurrent wheezing status for 8 children, and these cases were considered as missing data in the primary analysis. Overall, the population had multiple risk factors for wheezing, including high rates of asthma in the family, smokers in the home, and other children younger than 5 years of age in the home (Table 1).

At the first diet screen, 1 month after enrollment, 61.4% of those receiving maternal milk at randomization and 94.1% of those not receiving maternal milk were above the 200 IU/d threshold to transition to blinded study drug. At the 3-month adjusted age point, 90.2% of the infants receiving maternal milk at randomization and 100% of those not receiving maternal milk had been transitioned to blinded study drug. Feeding patterns are described in eTable 1 in Supplement 2.

Adherence was estimated for participants who received their final bottle of study drug and returned all bottles. Among participants who received their final bottle of study drug, rates of return of all bottles were 75 of 141 (53.2%) in the sustained group and 60 of 133 (45.1%) in the diet-limited group. Average estimated daily intake was 0.94 g in the sustained group and 0.91 g in the diet-limited group.

Recurrent wheezing was experienced by 42 of 135 (31.1%) in the sustained supplementation group compared with 56 of 134 (41.8%) in the diet-limited supplementation group in the primary unadjusted analysis (difference, −10.7% [95% CI, −27.4% to −2.9%]; relative risk, 0.66 [95% CI, 0.47 to 0.94]; P = .02) (Table 2). Missing data sensitivity analysis supports the primary finding, in that associated averaged P values across a range of imputation scenarios were consistently less than .047 (eMethods in Supplement 2).

A planned model of recurrent wheezing adjusting for randomization strata, time in study, GA, and the variables associated with recurrent wheezing in bivariate analyses was also run. In this model, recurrent wheezing was significantly decreased in the sustained supplementation group (relative risk, 0.62 [95% CI, 0.44 to 0.87]; P = .005) (eTable 2 in Supplement 2).

For secondary outcomes, values were determined for each infant included, so there were no missing outcomes in the analyses. No significant differences between groups were seen in total medically attended illnesses or markers of allergy or bone health (Table 2). The Asthma Predictive Index was positive for 15 of 140 (10.7%) and 19 of 137 (13.9%) infants in the sustained and diet-limited groups, respectively (difference, −3.2% [95% CI, −12.4% to 4.6%]). Emergency department visits for a respiratory cause were experienced by 75 of 141 (53.2%) in the sustained group and 69 of 137 (50.4%) in the diet-limited group (difference, 2.8% [95% CI, −13.2% to 11.9%]). There was also not a significant difference in respiratory hospitalizations. No participants had a tibial speed of sound measurement more than 2 SDs below the mean, based on previously published norms, at 12 months.²⁰ Cases of elevated calcium and alkaline phosphatase levels were generally transient and did not differ between treatment groups. In analyses by time point, a significant decrease was seen in the sustained group at some, but not all, points for wheezing, respiratory medication use, and hospitalization (eTable 3 in Supplement 2).

There were no statistically significant differences between the groups in the number of infants experiencing any adverse events or serious adverse events (Table 2). Upper and lower respiratory tract infections were among the most commonly reported adverse events. Upper respiratory infections were experienced by 84 of 153 infants (54.9%) in the sustained group and 83 of 147 infants (56.5%) in the diet-limited group (difference, −1.6% [95% CI, −17.1% to 7.0%]). Lower respiratory infections were experienced by 33 of 153 infants (21.6%) in the sustained group and 37 of 147 infants (25.2%) in the diet-limited group (difference, −3.6% [95% CI, −16.4% to 4.4%]). No adverse events were attributed to vitamin D treatment or deficiency. No infants were diagnosed as having rickets. Adverse event counts by organ system are presented in eTable 4 in Supplement 2.

Median achieved circulating 25(OH)D concentrations were greater than 30 ng/mL at follow-up (Figure 2 and eTable 5 in Supplement 2). Separation between groups was primarily seen in the infants who were not receiving maternal milk at randomization. However, there was no significant interaction between maternal milk at randomization and group (P = .20). There was also no significant interaction between sex and randomization group (P = .32).

Quiz Ref IDIn this randomized clinical trial of black preterm infants, sustained supplementation with vitamin D compared with diet-limited supplementation resulted in a reduced risk of recurrent wheezing by 12 months’ adjusted age. Neither group demonstrated superiority in terms of bone health or allergic disease.

Both supplementation strategies targeted a total daily intake between 200 and 1000 IU of vitamin D, generally within the high and low ranges recommended by the American Academy of Pediatrics and Institute of Medicine^18,19,25; infants consuming a high volume of formula designed for feeding preterm infants may have exceeded that range. Infant dietary vitamin D intake also tends to increase over time; therefore, total intake was not constant over the study period. Epidemiologic studies in a northern Finland cohort showed an association between infant supplementation and allergy and atopy later in life,²⁶ and other cohort studies have found increased wheezing or asthma in black infants being supplemented with up to 400 IU/d.^10,11Quiz Ref ID However, this study suggests a benefit of sustained supplementation with 400 IU/d with regard to recurrent wheezing, without an increase in allergy or eczema. The mechanisms of such an effect are unknown, but could potentially include developmental alterations in the airway or immune system, or acute alterations in inflammation or response to infection.

In pediatric and adult populations, several trials have failed to show a significant benefit of vitamin D supplementation on respiratory outcomes, including asthma or upper respiratory infections.^14,27-29 In addition, in a study of extremely preterm infants, vitamin D did not affect duration of respiratory support.³⁰ However, multiple observational studies have shown an association between respiratory disease and low vitamin D levels.^31-35 In term infants, differences in cord 25(OH)D levels were associated with meaningful differences in the risk of lower respiratory tract infection with respiratory syncytial virus.³² A recent individual patient data meta-analysis found that supplementation decreases respiratory infections, particularly in vitamin D–deficient patients or those receiving daily dosing.^36,37 However, results of studies in other populations should not be extrapolated to preterm infants who may be in a critical developmental window with regard to the effects of vitamin D on the pulmonary and immune systems.

This effectiveness study has several strengths, including targeting clinically relevant outcomes in a high-risk and under-studied population of preterm infants, with low rates of attrition during the study. Validated questions were used that have been shown to perform well in this population and were concurrently used in other trials.^12-15,38 Stratification of randomization by receipt of maternal milk intake in the prior 24 hours aimed to balance maternal milk exposure between randomization groups, due to the immunologic and dietary differences between breastfed and exclusively formula fed infants, including the potential protective role of maternal milk with regard to wheezing and infection.

This study has several limitations. First, despite the use of a validated questionnaire, there is potential for misclassification of wheezing by questionnaire. Second, the study was not powered to detect significant differences in secondary outcomes such as hospitalization or serious adverse events. Third, resource utilization, such as medically attended illnesses, are also highly driven by nonbiological social factors. Fourth, the significance level for a priori secondary outcomes was not adjusted for multiple comparisons. Therefore, statistically significant results in secondary outcomes, such as the analyses by follow-up point, should be interpreted as exploratory. Fifth, this effectiveness study tested 2 dosing regimens within the bounds of common supplementation recommendations in a population of black preterm infants born at 28^0/7 to 36^6/7 weeks’ GA with a paucity of major neonatal morbidities. The sustained strategy may or may not reflect the optimum dosing strategy in this population; the results cannot be generalized to support the efficacy or adverse event profiles of higher doses; nor can they be generalized to infants of other races, those born at less than 28 weeks’ GA, or infants with bronchopulmonary dysplasia.

Among black infants born preterm, sustained supplementation with vitamin D, compared with diet-limited supplementation, resulted in a reduced risk of recurrent wheeze by 12 months’ adjusted age. Future research is needed to better understand the mechanisms and longer-term effects of vitamin D supplementation on wheezing in children born preterm.

Corresponding Author: Anna Maria Hibbs, MD, MSCE, Rainbow Babies and Children’s Hospital, Division of Neonatology, 11100 Euclid Ave, Ste 3100, Cleveland, OH 44106 (annamaria.hibbs@cwru.edu).

Accepted for Publication: April 17, 2018.

Correction: This article was corrected on August 14, 2018, to include an undisclosed potential conflict of interest for Dr Ross.

Author Contributions: Drs Tatsuoka and Hibbs had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Hibbs, Ross, Zimmerman.

Acquisition, analysis, or interpretation of data: Hibbs, Ross, Kerns, Wagner, Groh-Wargo, Fuloria, Minich, Tatsuoka.

Drafting of the manuscript: Hibbs, Wagner, Fuloria, Tatsuoka.

Critical revision of the manuscript for important intellectual content: Hibbs, Ross, Kerns, Wagner, Groh-Wargo, Fuloria, Zimmerman, Minich.

Statistical analysis: Minich, Tatsuoka.

Obtained funding: Hibbs.

Administrative, technical, or material support: Hibbs, Wagner, Groh-Wargo, Fuloria, Zimmerman, Tatsuoka.

Supervision: Hibbs, Groh-Wargo, Tatsuoka.

Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Drs Hibbs, Kerns, and Zimmerman reported receiving grants from the National Heart, Lung, and Blood Institute (NHLBI) and Office of Dietary Supplements (ODS). Dr Ross reported receiving grants and/or nonfinancial support from the National Institutes of Health, ODS, Boehringer Ingelheim, Teva, GlaxoSmithKline, Merck, Flamel, and Jazz Pharmaceuticals, and Otsuka Pharma/Pharmavite. Dr Wagner reported receiving grants and/or personal fees from the National Institutes of Health/NHLBI and Church & Dwight Inc, for which she serves as a scientific consultant. Dr Groh-Wargo reported receiving speaker honoraria from Abbott Nutrition. Dr Tatsuoka reported receiving grants from the National Institutes of Health, ODS, National Science Foundation, and Biogen. No other disclosures were reported.

Funding/Support: This study was funded by the NHLBI and ODS (grant R01HL109293).

Role of the Funder/Sponsor: The NHLBI monitored the conduct of the trial and selected the membership of the data and safety monitoring board. Communications between the investigators and the data and safety monitoring board occurred through the staff of the NHLBI. The NHLBI and ODS had no role in the design and conduct of the study; collection, management, analysis, and interpretation of data; preparation, review or approval of the manuscript; or the decision to submit the manuscript for publication.

Additional Information: Clinical centers: University Hospitals Cleveland Medical Center: Anna Maria Hibbs (site principal investigator [PI]), Arlene Zadell, Leslie Clarke, Amy Zipp, Eileen Goldblatt, Matthew Wright, Allison Payne, Mary Nock, and Andrea Trembath. Medical University of South Carolina: Carol Wagner (site PI), Pam Smith, Katreia Dabner, and Lynn Smith. Montefiore Medical Center: Mamta Fuloria (site PI), Sonia Sheehan, Shreyans Bengani, Mohamed Farooq Ahamed, and Melissa Vega. MetroHealth Medical Center: Sharon Groh-Wargo (site PI), Diane Fierst, Julie Gualtier, and Rose Harcar-Sevcik.

Additional Contributions: We thank Lydia Furman, MD (Case Western Reserve University and Rainbow Babies and Children’s Hospital), for serving as the medical monitor for this study, Arlene Zadell, BSN (Rainbow Babies and Children’s Hospital), for serving as study coordinator, Jeremy Fondran (Case Western Reserve University) for serving as clinical data manager, and Denise Babineau, PhD (Case Western Reserve University; Rho Inc), for developing the statistical plan. Dr Furman was not compensated for her role. Mr Fondran, Ms Zadell, and Dr Babineau received salary support for their work. We thank the participants of D-Wheeze for their contributions to the trial.